Recommendation Based On Video-based Audience Sentiment
AssigneeZoom Video Communications, Inc.
InventorVi Dinh Chau
Life of this patent
Estimating the term…
Abstract
Video data from audience participants reacting to a speaker participation during a conference is obtained. The video data is processed to detect and recognize reactions based on a speaker presentation. Sentiment types are determined for the recognized reactions in view of a context of the speaker presentation. An engagement level is determined based on aggregated sentiment types for the audience participants. A real-time recommendation output is presented based on the engagement level. The real-time recommendation output provides suggestive actions for the speaker participant based on a positive or negative engagement level.
Claims20
A method, comprising:
- determining, by a server during a video conference, meanings of reactions of one or more audience participants to a speaker participant based on real-time transcription of the video conference;
- determining, by the server, sentiment types of the one or more audience participants based on the determined meanings of the reactions;
- determining, by the server, an engagement level based on most frequent counts of the determined sentiment types;
- aggregating, by the server using a machine learning system, engagement levels, real-time recommendations, and speaker participant behaviors over multiple conferences to determine a real-time recommendation;
- and
- providing, by the server using the machine learning system, an outputting of the real-time recommendation at a device associated with the speaker participant based on the real-time recommendations that were previously effective.
An apparatus, comprising:
- a memory;
- and
- a processor configured to execute instructions stored in the memory to:
- determine, during a video conference, meanings of reactions of one or more audience participants to a speaker participant based on real-time transcription of the video conference;
- determine sentiment types of the one or more audience participants based on the meanings of the reactions;
- determine an engagement level based on most frequent counts of the sentiment types;
- aggregate, using a machine learning system, engagement levels, real-time recommendations, and speaker participant behaviors over multiple conferences to determine a real-time recommendation;
- and
- provide, using the machine learning system, an outputting of a real-time recommendation at a device associated with the speaker participant based on the real-time recommendations that were previously effective.
A non-transitory computer readable medium storing instructions operable to cause one or more processors to perform operations comprising:
- determining, by a server during a video conference, meanings of reactions of one or more audience participants to a speaker participant based on real-time transcription of the video conference;
- determining, by a server, sentiment types of the one or more audience participants based on the meanings of the reactions;
- determining, by the server, an engagement level based on most frequent counts of the sentiment types;
- aggregating, by the server, engagement levels, real-time recommendations, and speaker participant behaviors over multiple conferences to determine a real-time recommendation;
- and
- providing, by the server using the machine learning system, an outputting of the real-time recommendation at a device associated with the speaker participant based on the real-time recommendations that were previously effective.
Dependencies are inferred from the claim text.